Comments
Many people have found value in this 80,000 Hours podcast episode from July 2026: Spencer Greenberg on staying sane while trying to save the world.
The highlights below are Julia's picks of the best bits, lightly reformatted, for those who prefer a written format. Thanks to Spencer, Luisa, and 80,000 Hours for allowing me to republish their material! Most quotes are from Spencer, and some are from Luisa.
Nutshell version at 80,000 Hours: 5 minute read
Medium version below: 17 minute read
Full podcast at 80,000 Hours: 2 hour listen, or read the full transcript
………….
Spencer Greenberg — founder of ClearerThinking.org and coauthor of The 12 Levers: The Complete Psychological Toolkit for Improving Your Life — talks with Luisa Rodriguez about the psychological toll of working on existential risk, AI safety, factory farming, and other high-stakes problems.
Drawing on a recent survey his team ran, he argues that grinding yourself down is self-defeating even on its own terms, and that the way out runs through understanding your actual psychology rather than forcing an idealised one. He and Luisa discuss practical, evidence-based tools from his new book, The 12 Levers, that can help with the most common challenges.
………….
At the top of the list are: therapy, medication, talking to friends and family about the issues.
We surveyed people who work on these high-impact cause areas like existential risk, animal welfare, and AI. We ended up getting about 60 to 110 respondents, depending on the question. A lot of the most instructive stuff was around when we asked the question, “What would you advise someone who’s very similar to you who’s getting into this field, in terms of dealing with these psychological challenges?”
Even if your goal were just to get the maximum performance, you have to be realistic about what kind of body you have and what kind of mind you have — and grinding all the way is not actually the way to get maximum performance. It’s a naive version of trying to get maximum performance.
We’re such limited beings, and that no matter how hard we work to be unlimited beings, we’re never going to get there. All of these things are just ideals that we strive for and we won’t come close.
For example, I really care about trying to be rational, and I regularly see how far I am from that goal. And you could let that kind of undercut your motivation, being like, “Oh man, I’m so far from that,” or you can turn that around and you can say, “Hey, I can keep becoming a better version of myself and move incrementally towards this more ideal form.” And that can be exciting. I don’t think it has to be demotivating. It can be motivating, actually: being limited, but seeing how you can progress and improve.
But I would also say that failing to accept your limitations can actually be disastrous towards the goals that you care about. So the more accurately you can see yourself — what you are, what your limitations are — the better you can be at all the things that you care about.
What is hypervigilance, and what’s the point of it? Effectively, it’s a state that’s preparing you to take drastic action. Like there’s something dangerous that’s coming: you’ve got to go, you’ve got to act. Unfortunately, it’s really not a healthy state to be in all the time. It’s not good for your body, it’s not good for your mind. It actually affects your digestion when you’re in a hypervigilant state.
I like to think about professional athletes here. Do professional athletes want to be stressing their bodies all the time? If you’re training to be in the Olympics, and all you care about is the goal of winning the Olympics, are you just stressing your body to the absolute limit? The answer is absolutely no, because that is not the performance-optimising move. The performance-optimising move involves genuine rest; it involves variation of things. You’re not always working just on that one skill set. You’re going to have off days. You’re going to have days where you’re practicing other things or doing cross training or whatever — and that’s even if you just care about optimising for the goal.”
When people feel like there’s a frightening thing in the world, they’ll often track it. You look at early days of the COVID pandemic, people were absolutely glued to any source of information, couldn’t talk about anything else. And I think it is that vigilant system of like, “This really scary thing is there. We’ve all got to monitor it really closely.”
You need a system for monitoring it. That is important if that’s your field, if that’s what you’re focused on. But once you have that in place, it actually can give you permission to not monitor other times
Let’s say you’re in the field of AI, and things are moving so fast. It may be even more difficult to deal with, just because there is information you really want to stay on top of and it is changing a lot. And there, I would say the same kind of strategy can be useful: being really aware of where am I getting my news, what are the good news sources, what’s the stuff that’s going to keep me up to date?
And then trying to block out all the noise. You don’t need to constantly scroll down and get every single new post on X about AI, a bunch of which are probably complete exaggerations anyway or bad takes. You know, figure out your trusted news sources and that’s where you’re going to get your information.
People go to social media as a form of avoidance or as a form of attempting to unwind, right? So they’re like, “I don’t want to think about this thing. I’m stuck on this thing. Let me open up social media.” Or like, “I’ve just been working hard, let me unwind.” And the irony is that they end up often in this sort of hypervigilant state when they’re on social media, where it’s like not even necessarily doing a good job of getting them to relax, or maybe they’ve avoided this one thing and now they’re stressed out about this other thing they’re looking at.
What else is restful? What else shares that same need? Maybe walking your dog or going for a walk in nature or having a call with a friend or whatever it is. And maybe it takes a little bit more planning to make that happen; it’s not quite as mindless — but if you could swap it out with something that serves the same need at least as well, hopefully better, but where afterwards you’re like, “I’m glad I did that; that was a good use of time,” it can not only serve our needs better, but actually make life more meaningful and fulfilling, and I think be more restorative.
You need to do meaningful work, that’s great, but you’re trying to replace your need to rest, and it doesn’t satisfy your need to rest. It’s a different thing. And actually, you can kind of grind yourself to a pulp and make it hard for you to function if you’re never letting yourself actually satisfy that rest need. Or if your “rest” is not actually restful, like you’re actually stressed out doing your rest because you’re scrolling stressful AI news or whatever it is.
Based on Spencer’s interview with Stephen Zerfas on the Clearer Thinking podcast, about motivation.
How are we motivating ourselves to get stuff done?
A lot of really high-achieving people use what [Zerfas] calls “dirty fuel,” which is essentially we beat ourselves up or are really harsh on ourselves as a way to get really good performance. And the irony of it is often it does work for people, but it has tremendous psychological costs, and also it can be unsustainable. It might actually be bad in the long run where they burn out or actually don’t achieve their goals because of the psychological costs. And then there’s this idea of “clean fuel,” which is other ways to motivate yourself that don’t have the psychological costs, or even have psychological benefits.
Being human, there are a lot of ways we can motivate ourselves. One of them is telling ourselves we’re never doing enough or feeling guilt, but we can have other forms of motivation. We could have social motivation: maybe I’ll collaborate with someone, and that will motivate me. We can have reward-based motivation: if I do this, I’ll give myself this cool reward. Et cetera, et cetera. There are many forms of motivation. I could try to develop a feeling of excitement around this project in particular. I don’t think there’s like a switch we can flip, but we can experiment in small ways on lower-stakes things on other ways to motivate ourselves
There can be a kind of boom-and-bust cycle of periods of giving work kind of everything you’ve got, sprinting really hard, but then neglecting your own needs and eventually crashing.
How would you want to act in these difficult situations? What would the ideal version of you do? Let’s say you think that there’s a really high chance that existential risk is on the horizon. Suppose you believe that. Would you want yourself to just be freaked out 100% of the time? Is that actually the ideal version of you? And what is the ideal version of you? How often would you be freaked out, and to what degree? And what would that fear do for you?
When you kind of do that thought experiment, you realise that actually just being freaked out all the time is like, on some level it feels maybe proportionate to you, but it’s actually not what you would want of yourself. And the ideal version of you wouldn’t be in that state.
Imagine you had a magic dial and you could control how much of different emotions you would feel. Let’s take your fear: where would you dial that dial?
[People] sometimes feel like therapists are trying to push them down to zero. Not that that’s achievable, but they might feel that, and they may actually have resistance to that because they might feel like zero is not the right level.
So I’d ask that of you: would you actually leave the dial exactly where it is?
[About guilt]: what is it detecting in the smoke detector model? It’s detecting a feeling like you’ve done wrong or you’ve violated a moral standard. And then I would ask the question: have you really? Like, I don’t know how much you work, but is working 80 hours required as a moral standard? Should everyone feel guilty for not working 80 hours if they could make a difference, like you believe you can, et cetera?
What is this guilt doing for you, and what’s its cost? On that magic dial, how much guilt would you choose to feel? Would you choose to feel the amount on the magic dial if you could control it? Is the guilt actually helping you do a better job, or is it actually impairing you and making you do a worse job?
My own kind of cycles of sprinting really hard, feeling like I have this sense of duty to do as much as I can, is related to feeling like I’m special; that I might be one of a few people who can make a difference to whether we solve these global problems.
So a bit of a savior complex. It feels kind of uncomfortably entangled with feelings of grandiosity and narcissism, which feels yucky saying aloud. But yeah, I find myself attached to the idea that me specifically, I might have a lot to contribute. It feels kind of related to having to main character energy.
And the fact that I’m really beating myself up for not eking out a few extra hours a week to work does just feel like it’s really coming from this “maybe I’m special and I’m going to make a difference.” Whereas really, realistically, I think the more correct thing is, “I am playing my part. My part is not enormous. Whether I work a few extra hours this week is not going to save the world or determine whether things go well or badly.”
The survey actually had kind of a striking pattern, where people said their anxiety went down once they started actually working on the problem — that doing something, even something small, was more bearable than just passively dreading. For some people, there’s something like agency that is actually really helpful for them.
Getting really pragmatic and just working on the behavior can often make people feel a lot better. And that tracks with “find a problem and go work on it,” even if you have this existential angst.
We’re talking about different probabilities of different outcomes — and I would be very surprised if there’s absolutely no chance that we succeed at our goals. I think there’s some chance at least that things could go well.
And I find it helpful to think about how I’m going to help in the set of worlds where this is a tractable problem. I’m not going to help in the set of worlds where it isn’t tractable and we’re doomed from the start. I’m also not going to help in the set of worlds where it isn’t a problem at all. But in the set of worlds where what we can do could make a difference, I’m going to play a part, and that’s really meaningful.
So I let go, and I feel sad about the worlds where I couldn’t have done anything at all to make a difference, but I act as if I’m in the worlds where I can — because that’s all I can do, really.
Do the coping strategies that help with individual death help with the extinction of humanity?
We really have to grapple with the enormity of the value loss that would actually be tremendously bad. And I think the way to cope with it is not to pretend it’s less bad than it is. But an interesting thing is that even things that are really, really bad, you can become more psychologically strong in the face of them.
Think about people going into battle for some just cause. Let’s say their people are being attacked and they have to defend their people. And you think of, for example, a courageous warrior going into battle: they might die, their whole family might be wiped out, their entire community and their whole civilisation might be wiped out — but it doesn’t mean that they have to go in in terror. And in fact, going into that battle in terror is dysfunctional: it actually makes it more likely that they’re going to get wiped out, right?
So I think we need to understand that the enormity of something being difficult does not imply you have to just be in terror all the time.
Different psychological strategies that are helpful for different people. One of them is going to be around courage, and this idea of being strong in the face of adversity and standing up to difficult things. For some people, that can actually be inspiring in a way.
For others, the best strategy will be acceptance. But acceptance does not mean that you’re not resisting it and trying to stop it. Acceptance is a psychological attitude of, “I accept that this thing is a possibility. My mind is not rebelling against reality, but that doesn’t mean I’m not going to work my hardest to fight it. In fact, I might be more effective fighting it if I accept it.”
A useful metaphor here is: imagine a patient that is likely to be terminally ill. Like, they have a high probability of being terminally ill, but there’s still some stuff they can do to increase the probability of survival. Let’s say they have a year to live, but it’s a chance to survive if they can do everything right, but they probably won’t no matter what they do.
What is the best attitude for the patient to have? There will be some individual variation, but I think one healthy attitude would be to try to fully accept the possibility that they’re going to die — so they’re no longer mentally rebelling against that while still taking effective action to reduce that probability.
The word “acceptance” is a little funny there, because we don’t mean “accept” as in, “OK, I’m fine with it; I’m not going to resist it.” We mean “accept” in that you’re no longer mentally rebelling against it. You’re no longer saying, “This can’t be happening. I won’t accept this. Fuck this.” You’re saying, “Here’s reality. And now I’m going to operate under the assumption of reality.”
We can actually still have really good relationships, even if we have these differences in important beliefs.
There’s ways to talk about it where you talk about it more in terms of your experience, rather than in the conversation requiring they buy into a set of assumptions.
With animals, it might be that you saw something that disturbed you. And you can say, “I saw this thing and I felt really disturbed by it.” That’s not making a factual claim about the objective nature of this problem, but it’s like, I was really bothered by this, and then they can relate to the fact that you are bothered and connect over that.
Or let’s say with AI, maybe your friends and family have other concerns about AI, and they’re not the same as yours. You could talk about it in a broader way. Say, “I’m really concerned about AI. In my job, I have to think about it all the time, and actually I find it disturbing to think about.” You don’t have to get into the exact facts and then negotiate what’s actually true. A lot of our friends and family, they want to connect with us and they want us to feel better and they want to empathise with us. But we can accidentally derail it by making it about the facts.
There’s incredible amounts of disagreement around these topics, so if you’re super duper confident you’ve got the correct answer, it’s worth being a little bit epistemically humble.
Very rarely are people willing to be convinced unless they could see the possibility that they could convince you too, right? If I come in, like, “Here’s why you’re wrong,” you’re not going to listen to me. But if I’m like, “Let me hear you out. I want to hear your perspective. Can I also tell you my perspective?” Now we can actually have a real dialogue.
Whether to have kids, work/life balance, personal spending versus donating, saving for retirement, doing a job you find difficult or unenjoyable, where to live, risk tolerance in career choices. It's kind of just one question over and over again: how much of your life are you allowed to keep for yourself versus do you want to spend on solving global problems?
You probably really care about sentient beings not suffering and the world not ending. Those are probably deep, important things that connect to your intrinsic values. And you also care about connection to your loved ones and maintaining your relationship with them, and having pleasure in your life, and not suffering yourself too much.
So then we can now ask the empirical question: how much do you care about each of those? And the answer to that is not going to be a stable number. In my experience, just as a claim about human psychology, that is going to depend on to what extent you’re neglecting each of them. If you are investing all your focus on reducing suffering in the world, there’s a good chance that the other values will pull a lot harder. And if you really inspect, you’re going to say, “These are mattering more now to me because I neglected them. I’ve not been satisfying these values.”
Then the way I approach it is I try to strike a balance between these things. I try to not neglect my intrinsic values, but let them sort of come in communication with each other and find a balance between them that feels right. And that’s a psychological quest. It’s not a quest about the nature of philosophy or the universe.
When someone is all-in on the mission, and only thinks about the mission and doesn’t take into account their other values: they start to pull harder on them psychologically, and I think that leads to things like burnout. I think one of the causes of burnout is neglecting your other important values.
I think you’ll pay a very significant price to not understand your values.
[More in the episode about Spencer’s approach: valuism]
[On various approaches to understanding what different parts of you want]
The classic general technique is you want to notice that there’s some part of you that wants something or has urges or needs not met. You can give it a name. It can be helpful to give it a name, because then you can talk about it or refer to it later. So maybe you’ll say, like, “I’m going to call it the Critic. It’s the part of me that tells me I’m never doing enough, I’m never good enough.”
And then you want to put yourself into a state where you can talk to it very compassionately, without being reactive. And they give different names to this, like the Self or the Compassionate Self or the Adult or whatever it is. But you’re kind of going to be dialoguing with this part from a very compassionate, healthy state, and you’re going to try to just explore what that part needs and wants and cares about, and try to help it have its needs met.
I had to kind of grieve the part of me that wanted to work 80-hour weeks because it thought that was the most morally good thing to do in order to fulfil the wishes of the part of me that really wanted to be a mom. And I think that part of me that wanted to work harder would have just kept being like, “You’re a bad person! You’re a bad person!” unless I’d really heard it out and been like, “I hear you. Thank you. You want me to be a good person, I appreciate that. But I’ve decided, full-eyed looking honestly at things, that I want to be a mom anyways.”
..............
Full episode at 80,000 Hours.
Spencer's new book, The 12 Levers, teaches related psychological tools in concrete terms.
